This conversation examines productive inference and infrastructure efficiency for artificial intelligence. Betsy Chernoff of WEKA appears in a discussion hosted by John Furrier of theCUBE and produced by SiliconANGLE. Chernoff outlines storage-driven approaches to inference and the shift from training to inference workloads; they explain benchmarking methodologies, agentic AI traces and key-value cache management, and emphasize that storage, CPUs and networking must operate together with graphics processing units to deliver production-grade AI services and measurable intelligence.
Chernoff highlights the need for production-grade benchmarks and new metrics to assess productive inference, including time to first token, time to last token, tokens per watt per second and a metric referred to as "good-put." They describe Augmented Memory Grid with NeuralMesh as a key-value cache offload that serves context at memory or dynamic random-access memory speed, enabling WEKA to demonstrate a reported 2.4x increase in requests. The discussion also cites a scorecard by Steve McDowell and the buyer guide from 451 Research as frameworks for procurement and validation.
Watch to learn more about inference benchmarking, KV cache strategies, storage-driven architectures, token economics and system-level performance for production AI.
Almost there!
We just sent you a verification email. Please verify your account to gain access to
. If you don’t think you received an email check your
spam folder.
In order to sign in, enter the email address you used to registered for the event. Once completed, you will receive an email with a verification link. Open this link to automatically sign into the site.
Register For Cube365 Events Platform
Please fill out the information below. You will recieve an email with a verification link confirming your registration. Click the link to automatically sign into the site.
I want my badge and interests to be visible to all attendees.
Checking this box will display your presense on the attendees list, view your profile and allow other attendees to contact you via 1-1 chat. Read the Privacy Policy. At any time, you can choose to disable this preference.
Select your Interests!
add
Upload your photo
Uploading..
You are already logged into TheCUBE Network as
Share
Share clip
The clip will play by default when shared. The entire video and transcript is also accessible via the link.
This conversation examines productive inference and infrastructure efficiency for artificial intelligence. Betsy Chernoff of WEKA appears in a discussion hosted by John Furrier of theCUBE and produced by SiliconANGLE. Chernoff outlines storage-driven approaches to inference and the shift from training to inference workloads; they explain benchmarking methodologies, agentic AI traces and key-value cache management, and emphasize that storage, CPUs and networking must operate together with graphics processing units to deliver production-grade AI services and me...Read more